Although it is logical to assume that all eukaryotes share a common mechanism for pre-mRNA splicing, gene transfer experiments indicate that critical differences exist between the events leading to mammalian, yeast and plant intron recognition. Differences which also exist between plant introns have prevented in vivo expression of some plant genes containing introns in transgenic plants. In vivo analysis of the cis-acting factors required for intron recognition in plant nuclei has allowed us to develop a model for intron recognition stating that AU elements spread throughout the length of plant introns roughly define intron boundaries by generating strong AU-transition points and masking internal cryptic sites. Potential splice sites are then selected in a position-dependent manner if they are located upstream (5' splice site) or downstream (3' splice site) from these AU transition points and not if they are embedded within AU-rich intron sequences. This mode of recognition relaxes the need for strong splice site and branchpoint consensus sequences and suggests that plant splicing machineries rely on a variety of novel trans-acting factors. The prominence of AU-rich introns and AU transition points at the intron/exon boundaries of Tetrahymena, Drosophila, C. elegans and S. pombe introns suggests that similar mechanisms for intron recognition and splice site definition may operate in a variety of species. Experiments proposed here are aimed at fully defining cis-acting sequences mediating intron recognition in plant nuclei and isolating trans-acting factors that interact with these sequences. Objectives are: 1) To test the universality of this model for intron recognition on other introns; 2) To define in detail cis-acting elements responsible in vivo for 5' and 3' splice site selection in AU-rich introns; 3) To identify trans-acting factors that interact with the cis-acting sequences, especially AU-binding proteins that appear to differentiate introns from exons in vivo and other proteins that enhance recognition of plant introns in an in vitro HeLa-plant complementation system. These experiments will identify splicing factors and recognition schemes that may mediate excision of AU-rich introns in many species and will determine critical associations that commit pre-mRNAs of this type of a splicing pathway. Ultimately, this information should clarify similarities and differences between the modes of intron recognition operating in mammalian, yeast and plant nuclei and provide for a more thorough mechanistic definition of pre-mRNA splicing.